1902] A MORPHOLOGICAL STUDY OF ASCLEPIADACEAE 407 



\ 



antipodals may still be seen flattened out against the sac -walls 

 when there are approximately thirty-two endosperm nuclei 

 (y?^. 5^). Frequently in this stage one of the synergids could 

 still be recognized as a small, poorly staining nucleus, crowded 

 into the space at the base of the ^g<g {^fig^ 5^)- 



The ^^^ after fertilization forms a distinct wall and rests. No 

 division occurs before the endosperm has passed its sixteen-celled 

 stage {^fig^ 5/). This leads one to question whether the reported 

 cases of endosperm division before fertilization, e, g,, Casuarina 

 (15), may not be due to misinterpretation, the resting tgg being 

 mistaken for an unfertilized one. 



Some changes of interest occur in the ovule after fertiliza- 

 tion. Before fertilization the placenta begins the formation of a 

 longitudinal ridge under each row of funiculi. This continues 

 to grow higher until when the seeds mature it is a membrane 

 about 3 ^^ high. In it can be seen plainly with the naked eye the 

 bundles passing to the ovules. On account of the growth of the 

 lateral portions and ends of the ovules, and possibly pressure 

 from the ovary wall, funiculo-micropylar flattening takes place. 

 It seems that growth along the sides has most to do with the 

 flattening, for the tissue in the edges is quite meristematic. 

 About the time of the third division of the endosperm the epi- 

 dermis at the micropylar end of the ovule — the one nearest the 

 top of the ovary — becomes intensely active. The cells elon- 

 gate enormously and form the pappus. The hairs composing it 

 remain single cells with a single nucleus in each. Chauveaud (3) 

 reports two nuclei in each in Cynanchum. The pappus is tightly 

 packed in the grooves between the placental ridges, and the free 

 ends extend towards the apex of the pod. The micropyle in A, 

 Cornuti and A. Stdlivantii is still visible with a hand lens when 

 the seed begins to turn brown. The funiculus never thickens 

 much, so that in its later stages it is little more than a vascular 



bundle and is very easily broken off. 



The changes resulting in the formation of the pod also 

 deserve attention. At or shortly before the time of fertilization, 

 the outer epidermis becomes active, as is shown by its deep 

 stain and protoplasmic contents. The surface usually becomes 



